Munich Personal RePEc Archive Identification and Analysis of the Evolution of Local Productive Arrangements Welter, Caroline and Cypriano, Luiz and Centurião, Daniel Western Paraná State University, Western Paraná State University, Western Paraná State University 10 August 2021 Online at https://mpra.ub.uni-muenchen.de/109222/ MPRA Paper No. 109222, posted 21 Aug 2021 12:02 UTC Journal of Regional Analysis & Policy 51(2): ??—?? Identification and Analysis of the Evolution of Productive Arrangements Caroline Welter Western Paran´aState University Luiz Cypriano Western Paran´aState University Daniel Centuri˜ao Western Paran´aState University Received: 03/30/2021 Accepted: 08/10/2021 Abstract This study sought to identify and analyze the evolution of Local Productive Arrangements (LPAs) in the sectors of extractive and manufacturing industry in the state of Paran´a, from 2006 to 2016. In the methodology, the normalized Concentration Index (nCI) and the Exploratory Spatial Data Analysis (ESDA), which consisted of the application of two statistical tests, the Global Moran’sI and the Local Moran’s I. With this methodological procedure it was possible to identify 57 industrial LPAs in the state of Paran´ain 23 economic sectors, with its greatest incidence in the Metropolitan Region of Curitiba and North Central. We conclude that there is a consolidation process with positive evolution of LPAs in the analyzed period, where the regions of Londrina and Curitiba and its surroundings stand out, as well as the activities of food and beverage production; textiles and clothing; wood and furniture; and the production of machinery, equipment, rubber, and plastics. 1 Introduction The Local Productive Arrangement (LPA) is a type of agglomeration of companies disseminated in Brazil and formalized by the Research Network on Local Productive and Innovative Systems (RedeSist), from the Economics Institute of the Federal University of Rio de Janeiro (IE/UFRJ). The concept of LPA can be considered the most appropriate for the Brazilian reality, as it involves not only the locational and sectorial proximity of companies, but also considers the interdependence between political, economic, and social actors, in order to strengthen the development of the place where the LPA is installed Cassiolato and Lastres (2003a). For Santos et al. (2004), LPA became relevant in terms of industrial and regional policy due to the competitive advantage it provides to companies. On the one hand, it is because location is a source of competitive advantages, regardless of tax incentives, transportation costs, and ease of access to raw materials or end consumers, and, on the other hand, these competitive advantages are related to the innovative capacity and diffusion knowledge, encouraging the entry of specialized services in the area, attracting investments and making good use of the regional image. The federal government used the LPAs as a measure to stimulate regional economic development, in order to combine the reactivation of agglomeration economies to support the regions, as pointed out by Fuini (2013). Public policies were put in place to support LPAs and methodologies were created to identify them throughout the Brazilian territory, in addition to the constitution of state nuclei, with the purpose of 67 Journal of Regional Analysis & Policy 51(2): ??—?? organizing governance at the regional level, for the elaboration and implementation of state policies (BNDES, 2009). In the state of Paran´a,following on from the initiative of the state government, 18 industrial LPAs were identified (IPARDES, 2009), of which, in 2008, 68,670 thousand jobs were created, representing a positive variation of 33.5% in relation to the first identification of LPAs occurred in 2003 by IPARDES and SEPL (2005a). Several identification initiatives have been carried out by researchers and institutional actors, such as Puga (2003), Suzigan et al. (2003a), IPARDES and SEPL (2005a, 2005b, 2005c), IPEA (2006), IPARDES (2009), and Brito et al. (2010), who applied LPAs identification methodologies, using indicators of productive specialization and employment variables and number of companies. However, these approaches, to some extent, did not address the evolution of these LPAs nor did they show their consolidation in specific activities or regions. In addition, the numerous studies of LAPs identification seen in the literature had a break in 2008, with much more modest approaches to identification. These facts motivated the choice of Paran´afor the analysis proposed in the paper. It was also verified the frequent use of political borders only from municipalities or microregions. Nev- ertheless, it is worth mentioning that political borders do not always reflect the location where the LPAs are present, as pointed out by Porter (1999), and they may geographically cover two microregions, three municipalities only, and so on. The use of the Locational Quotient (LQ) specialization index also has limitations, since it can overestimate the weight of a sector when the productive structure of the region under analysis is little diversified, and can underestimate the weight of a sector when the productive structure is very diverse, mainly in developed regions Fingleton et al. (2004). The criteria and filters are used to eliminate this limitation, but they are not free of criticism, since they are arbitrated by the researcher. All things considered, this work sought to verify the evolution of industrial LPAs in the state of Paran´a. Therefore, the objective was to identify LPAs in the extractive and manufacturing industry in the period from 2006 to 2016, and whether they grew in terms of the number of participating companies and the expansion of their activities, covering neighboring companies and municipalities. For this purpose, a methodology for identifying and mapping LPAs was used, which aims to overcome the limitations mentioned regarding the spatialization of LPAs, beyond geographical boundaries, and to capture other economic dimensions, in addition to the LQ usually used. The contributions of this research reside in updating the methodology of LPAs identification used in the literature, considering the effects that go beyond the geographical and political limits established between the spatial units and considering the diversity of activities that compose the industrial sector, besides minimizing the effect of using arbitrary filters in the identification of agglomerations. Additionally, the study also updates the works of LPAs identification that were developed in past periods, especially for Paran´a,and this temporal update allows the analysis of the behavior of LPAs in a time of extensive economic crisis, where the Brazilian industry suffered greatly, especially from the year 2011, according to Oreiro (2017). 2 Theoretical Framework About regional development, regionalization and clustering, the literature presents different approaches, Cavalcante (2008) highlights those who work with an evolutionary perspective, in an analysis of Regional Innovation Systems and Local Productive Arrangements (LPAs); Ehrnberg and Jacobsson (1997) and Cas- siolato and Szapiro (2002) authors who emphasize the analysis of industrial districts in a context of flexible specialization; Lipietz and Leborgne (1988) and Markusen et al. (1995) , authors who emphasize “innovative environments”, such as the European Research Group studies on innovative, Maillat (1995) those dealing with regionalization of theories of industrial organization; Storper (1994) author who emphasize business management and the formation of business clusters; and, finally, authors who consider increasing returns to scale and their effects on spatial phenomena, such as Krugman (1997) and Fujita et al. (1999). In the Brazilian reality, LPAs emerged as a theoretical effort to understand the productive agglomerations of small companies, as pointed out by Fuini (2013). The term LPA is theoretically based on the industrial 68 Journal of Regional Analysis & Policy 51(2): ??—?? districts of ? and the clusters of Porter (1999). Europe also has experiences of analysis using these concepts, as in the case of France and Italy Becattini (1999). However, they still lack a consensus on what characteristics define it. The term LPA was systematized in Brazil with the studies of RedeSist. Among the main researchers are Jos´eEduardo Cassiolato and Helena Maria Martins Lastres, who started their studies focused on the innovative means and national and local innovation systems from the 1990s. LPAs are constituted by the geographical concentration of companies, institutions, and organizations that have links with each other, even if incipient and that exercise their activities in specific economic sectors. Generally, there is an interaction between these actors, be it in the supply of inputs or services, training of human capital, promotion of RD, and design of public policies Cassiolato and Lastres (2003b). Fuini (2013), after analyzing the specialized literature, with a wide range of researchers in the 1990s and 2000s, verified some fundamental factors to define LPAs, namely, the territorialization of production, external economies, locational competitive advantages, innovation, and cooperation. For Santos et al. (2004, p.162) “The literature seems to be converging more and more towards the definition of LPAs as a geographical concentration of companies and other institutions that are related in a particular sector.”4.Nonetheless, they show that the definition of LPAs must be carried out based